OCT-GPA VERSUS VISUAL FIELD GPA – A

PROSPECTIVE COMPARISON OF STRUCTURAL AND FUNCTIONAL

PROGRESSION ANALYSIS IN GLAUCOMA – D. Chiselita, C. DanielescuIasi University of Medicine and Pharmacy, Romania

Purpose: To analyze the progression of visual field defects (VF) and/or retinal nerve fiber

layer (RNFL) thinning in glaucoma suspects and early glaucoma patients, using commercially

available software.

Methods: Prospective study that included 100 eyes of 50 patients (primary open angle glaucoma

patients or glaucoma suspects, as defined by the EGS Guidelines (1).

All eyes were subject to at least 4 reliable VF examinations (SITA Standard 24-2, Humphrey Field

Analyzer II, Carl Zeiss Meditec) and 4 RNFL measurements (Fast RNFL, Stratus OCT, Carl Zeiss

Meditec).

The VF progression was assessed using the Glaucoma Progression Analysis software (that

provides event analysis using at least 3 VF and trend analysis from at least 5 VF). The RNFL thinning

was analyzed with the Guided Progression Analysis (software version 5.0), a linear regression

analysis of average RNFL thickness that uses at least 4 measurements.

Results: The mean follow-up period was 39.93 ± 10.59 months. A total of 502 OCT scans and 543

visual fields were analyzed.

There were 64 glaucoma suspect eyes and 36 glaucomatous eyes. Only 6 eyes had moderate VF

loss (Hodapp criteria).

The mean number of RNFL measurements per eye was 5.02 ± 1.18 (4 to 8 measurements)

The mean number of visual field tests was 5.43 ± 1.24 (4 to 14 tests).

The annual rate of change for the mean RNFL thickness was -0.63 ± 2.65 µm/year (-0.03 ± 1.87

µm/year for glaucoma suspects and -1.36 ± 2.98 µm/year for glaucomatous eyes). The difference was

statistically significant (p=0,02). (Figure 1)

4

3

2

1

0

-1

-2

-3

-4

-5

Rate_change_OCT

Glaucom=0

Rate_change_OCT

Glaucom=1

Figure 1. The annual rate of change for the mean RNFL thickness for glaucoma suspects and glaucomatous eyes.

A trend analysis of visual fields was possible only in 51 eyes that had at least 5 reliable tests, which

enabled us to calculate the MD slope. There was no difference in MD slopes between glaucoma suspects

(0.19 ± 0.55 dB/year) and glaucomatous eyes (0.18 ± 0.41 dB/year). (Figure 2). The small positive tendance of

MD may be attributable to the learning effect.

1,5

1,0

0,5

0,0

-0,5

-1,0

-1,5

MD_slopeGlaucom=0

MD_slopeGlaucom=1

Figure 2. The MD slope for glaucoma suspects and glaucomatous eyes.

Two eyes (2%) were labeled as possibly progressing by the GPA software of the Stratus OCT.

In other two cases, a suspected progression after 4 exams was not confirmed afterwards.

(3 “Possible progression” and one “Likely progression”).

There was no visual field progression noticed by the trend analysis of MD slope.

There was no agreement between the structural and functional progression in the study group.

We present an interesting case that was considered as progressing by the GPA-OCT software

after 6 tests, but after subsequent testing the standard deviation became too high and the

progression was not confirmed. (Figure 3, 4)

The other eye of that patient was considered “Possible progression” by the visual field GPA

software. (Figure 4)

The visual field event analysis provided by the GPA software of HFA II has labeled 4 eyes (4%)

as progressing

Figure 3. GPA progression after 6 tests

Figure 4. The progression was not confirmed after subsequent testing

Figure 5. The other eye of the patient was labeled “Possible progression” by the visual field GPA software

Discussion: To our knowledge, in the literature there are no other comparisons between OCT – GPA and visual field – GPA.The difference in RNFL loss between groups was statistically, but not clinically significant. A possible explanation is the fact that glaucomatous eyes were treated, but suspect

eyes were not.In a comparison between GPA-OCT and visual field index (VFI), Leung found 18.1% eyes with average RNFL thinning and 14.65% eyes with VFI progression after at least 3 years of

follow-up. (2) However, their cohort comprised only glaucoma patients, whereas our cohort had a large proportion of glaucoma suspects.In early glaucoma patients we would expect structural progression to occur more frequently than functional progression. In our cohort the OCT-GPA found a very small proportion

of eyes as progressing.We found a possible explainaition in a paper by Lee et al, comparing trend analysis of RNFL-OCT with progression on red-free photographs: when the rate of global change of

RNFL was taken into account, the agreement with the photographs (as shown by the AUROC) was 0,68. The best Stratus parameter was a trend analysis performed on clock hours, with AUROC=0.78! (3)

Also, when taking into account the rate of average RNFL thickness progression, Cirrus HD-OCT outperformed the Stratus OCT in detecting more eyes with progression. (4)

Conclusions: In our cohort of glaucoma suspects and patients with early glaucoma, OCT GPA (a trend analysis) was not a very effective tool for detecting progression. We are expecting further software development, taking into account local thinning of RNFL as a more sensitive mark of glaucoma progression.

References

rd1. European Glaucoma Society - Terminology and Guidelines for Glaucoma, 3 Edition2. Leung CK, Cheung CY, Weinreb RN, Qiu K, Liu S, Li H, Xu G, Fan N, Pang CP, Tse KK, Lam DS Evaluation of retinal nerve fiber layer progression in glaucoma: a study on optical coherence tomography guided progression analysis. Invest Ophthalmol Vis Sci. 2010 Jan;51(1):217-223. Lee EJ, Kim TW, Weinreb RN, Park KH, Kim SH, Kim DM Trend-based analysis of retinal nerve fiber layer thickness measured by optical coherence tomography in eyes with localized nerve fiber layer defects. Invest Ophthalmol Vis Sci. 2011 Feb 28;52(2):1138-44. 4. Leung CK, Chiu V, Weinreb RN, Liu S, Ye C, Yu M, Cheung CY, Lai G, Lam DS.Evaluation of retinal nerve fiber layer progression in glaucoma: a comparison between spectral-domain and time-domain optical coherence tomography. Ophthalmology. 2011 Aug;118(8):1558-62.